Beyond Slower Stargardt Lesion Growth: Phase 3 Biomarker Data Further Differentiate Tinlarebant

For Stargardt disease, which currently has no approved treatment, slowing retinal degeneration is closer to the current therapeutic goal than improving vision in the short term. Belite Bio announced a new analysis from the Phase 3 DRAGON trial: after 25 months of treatment, quantitative autofluorescence (qAF) decreased by approximately 2% from baseline in adolescents receiving oral tinlarebant, compared with an increase of approximately 20% in the placebo group, adding a piece of mechanism-related evidence to the previously reported slowing of lesion growth.

qAF can be used to estimate the accumulation of toxic bisretinoid substances in the retina. These substances accumulate in the retinal pigment epithelium and are associated with progressive damage to photoreceptors and supporting tissues. By reducing retinol-binding protein 4-mediated vitamin A transport, tinlarebant is intended to reduce the formation of related toxic metabolites. Therefore, although the divergence in qAF trends between the two groups cannot be equated with patients seeing more clearly, it is consistent with the drug’s expected direction of action.

DRAGON, registered as NCT05244304, enrolled 104 adolescents with Stargardt disease across 11 jurisdictions and randomly assigned them in a 2:1 ratio to receive tinlarebant or placebo. The company previously said the trial met its primary endpoint: as measured by “definitely decreased autofluorescence” imaging, the annualized retinal lesion growth rate was 35.7% lower in the treatment group than in the placebo group. The new qAF results were presented orally on July 18, 2026, at the American Society of Retina Specialists annual meeting in Montreal.

Whether the structural imaging improvement can translate into functional benefits that patients can perceive remains a key question. External reports on the meeting data indicated that changes in vision over 24 months were small in both groups. This may reflect the slow progression of the disease and the trial’s still-short duration, and it also means that the lesion and qAF data alone cannot currently establish how much long-term vision the drug may preserve.

Regarding safety, the company said tinlarebant was generally well tolerated. Media accounts of the meeting presentation indicated that six serious adverse events occurred during the trial, four of them in the placebo group. None were ocular events, and none were explicitly described as treatment-related. Ocular reactions consistent with the drug’s pharmacologic effects—including delayed dark adaptation, xanthopsia, and impaired night vision—did occur. Most were reportedly mild and resolved while participants continued treatment; the full incidence rates and longer-term effects still require confirmation through formal review data.

Belite Bio said it has completed submission of the New Drug Application to the U.S. FDA, and tinlarebant has received Breakthrough Therapy and Orphan Drug designations. However, completion of the submission does not mean that the FDA has accepted the application, much less approved the drug for marketing. As of the publication of the relevant reports, the company was still awaiting the notice commonly known as the “Day 74 letter” to confirm whether the application had been accepted, the review timeline, and a possible Prescription Drug User Fee Act target date. Whether this therapy can become one of the first disease-modifying treatments for Stargardt disease will next depend on how regulators weigh imaging endpoints, functional benefits, and long-term safety.